Reasoning on BigBench Extra Hard
14.3mean@4ReSyn
Evaluation Results
| Method | Links | |
|---|---|---|
| ReSyn2026.02 | 14.3 | |
| Majority2026.02 | 13.1 | |
| Qwen 2.5-Instruct2026.02 | 11.2 | |
| Chance2026.02 | 8.83 |
| Method | Links | |
|---|---|---|
| ReSyn2026.02 | 14.3 | |
| Majority2026.02 | 13.1 | |
| Qwen 2.5-Instruct2026.02 | 11.2 | |
| Chance2026.02 | 8.83 |